An experimental and numerical study of solid/liquid phase change of water saturated porous media in the presence of natural convection
Abstract
An experimental and numerical study of the solidification and melting of porous media in rectangular cavities, cooled (or heated) from the side, in the presence of natural convection has been undertaken. Water was used as the phase change material. Different diameter glass beads or aluminum balls constituted the porous matrix. The purpose of the study was to obtain a fundamental understanding of the heat transfer processes and the effects of different parameters governing the phase change processes. The effects of different size beads, the imposed wall temperature to initiate and sustain the phase change process, the amount of superheating for freezing and the amount of subcooling for melting and the density inversion of water on the phase change were investigated experimentally. Quantitative information about the temperature distribution in the solid and liquid regions was obtained. The location of the fusion front was deduced from these measured temperature distributions. For both freezing and melting, the fluid flow, the shape and the velocity of the interface were found to depend on the temperature difference across the liquid region and the bead size (permeability of the porous medium). During the freezing of water-aluminum system, due to considerable supercooling, the fusion front could not be located from temperature measurements. Both freezing and melting processes were faster with aluminum balls than with glass beads, due to the higher effective thermal conductivity of the porous media with aluminum balls. The measured temperature distributions were compared with the predictions of a mathematical model that considered both conduction in the solid, thermal convection in the liquid, and the density inversion of water. This model is based on volumetric averaging of the macroscopic transport equations, with phase change assumed to occur volumetrically over a small temperature range. Good agreement was found between the experimental data and the numerical predictions for a system with glass beads. With aluminum balls, the model predicted higher melting rates. This is believed to be due to the uncertainty involved in predicting the effective thermal conductivity.
Degree
Ph.D.
Advisors
Viskanta, Purdue University.
Subject Area
Mechanical engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.